Cross Validate Plots
Kevin Lu
October 25, 2017
1. Source Cross Validate Functions
source("./Mean Reversion/RMR.010 Cross Validate Functions.R") 2. Load Data
pricing_data <- read_csv("./Mean Reversion/Raw Data/pricing data.csv", col_types = c("iTdddddddci")) 3. Set Parameters
Arguments
pricing_data: A dataframe containing pricing data from Poloneix gathered in tidy format.
time_resolution: The number of seconds that each observation spans. Takes values 300, 900, 1800, 7200, 14400, and 86400.
cutoff_date: A data representing the cutoff date between the train and test sets.
train_window: A period object from lubridate representing the length of time the train set covers.
test_window: A period object from lubridate representing the length of time the the test set covers. quote_currency: A string indicating the quote currency of the currency pairs. Can take values USDT or BTC.
adf_threshold: The threshold for the ADF test statistic. Pairs below this threshold are selected.
rolling_window: The number of observations used in each iteration of a rolling linear regression.
stop_threshold: A threshold for the spread z-score beyond which the strategy stops trading the coin pair.
signal_logic: A string indicating which logic to use to generate signals.
model_type: A string indicating whether raw prices or log prices should be used. Takes value “raw” or “log”.
number_pairs: The number of pairs to generate plots for.
time_resolution <- 300
train_window <- days(16)
test_window <- days(10)
quote_currency <- "USDT"
adf_threshold <- -2.96
# rolling_window <- 86400 / time_resolution * as.numeric(days(2)) / 86400
rolling_window <- 4608
stop_threshold <- 3.51
signal_logic <- "scaled"
model_type <- "raw"
number_pairs <- 8 4. Cross Validation September 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-09-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 14 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_DASH USDT_ZEC -5.380957
## 2 USDT_ZEC USDT_DASH -5.304666
## 3 USDT_ZEC USDT_XMR -3.961555
## 4 USDT_XMR USDT_ZEC -3.840609
## 5 USDT_REP USDT_ZEC -3.768945
## 6 USDT_DASH USDT_XMR -3.713431
## 7 USDT_REP USDT_XMR -3.666970
## 8 USDT_ZEC USDT_REP -3.597644
## 9 USDT_REP USDT_DASH -3.591031
## 10 USDT_DASH USDT_REP -3.533729
## 11 USDT_XMR USDT_DASH -3.471097
## 12 USDT_ETH USDT_LTC -3.415202
## 13 USDT_XMR USDT_REP -3.358202
## 14 USDT_LTC USDT_ETH -3.158724
5. Cross Validation August 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-08-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 15 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_REP USDT_ZEC -6.427602
## 2 USDT_ZEC USDT_REP -6.361551
## 3 USDT_LTC USDT_REP -6.019389
## 4 USDT_REP USDT_LTC -5.887493
## 5 USDT_ETH USDT_ZEC -5.275862
## 6 USDT_ZEC USDT_ETH -5.239483
## 7 USDT_LTC USDT_ZEC -4.881401
## 8 USDT_ZEC USDT_LTC -4.615123
## 9 USDT_REP USDT_ETH -4.400386
## 10 USDT_ETH USDT_REP -4.343464
## 11 USDT_XMR USDT_DASH -3.633904
## 12 USDT_DASH USDT_XMR -3.622352
## 13 USDT_LTC USDT_ETH -3.587846
## 14 USDT_ETH USDT_LTC -3.279832
## 15 USDT_REP USDT_DASH -3.074027
6. Cross Validation July 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-07-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 30 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_REP USDT_XMR -7.754601
## 2 USDT_XMR USDT_REP -7.547141
## 3 USDT_REP USDT_ZEC -5.773214
## 4 USDT_ZEC USDT_REP -5.378645
## 5 USDT_REP USDT_ETH -5.200190
## 6 USDT_DASH USDT_LTC -4.827813
## 7 USDT_ETH USDT_REP -4.778907
## 8 USDT_LTC USDT_DASH -4.602047
## 9 USDT_XMR USDT_ZEC -4.596609
## 10 USDT_BTC USDT_REP -4.537611
## # ... with 20 more rows
7. Cross Validation June 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-06-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 25 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_REP USDT_XMR -6.864618
## 2 USDT_XMR USDT_REP -6.776243
## 3 USDT_REP USDT_DASH -6.366745
## 4 USDT_DASH USDT_REP -6.201406
## 5 USDT_XMR USDT_DASH -6.157884
## 6 USDT_DASH USDT_XMR -6.083495
## 7 USDT_REP USDT_ZEC -5.143941
## 8 USDT_DASH USDT_ZEC -4.905823
## 9 USDT_ZEC USDT_REP -4.898359
## 10 USDT_ZEC USDT_DASH -4.867936
## # ... with 15 more rows
8. Cross Validation May 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-05-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 19 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_XMR USDT_ZEC -5.470549
## 2 USDT_DASH USDT_ZEC -5.451411
## 3 USDT_XMR USDT_DASH -5.406064
## 4 USDT_ZEC USDT_DASH -5.365351
## 5 USDT_DASH USDT_XMR -5.359242
## 6 USDT_ZEC USDT_XMR -5.339887
## 7 USDT_XMR USDT_ETH -4.744628
## 8 USDT_ETH USDT_XMR -4.330518
## 9 USDT_ZEC USDT_ETH -4.093348
## 10 USDT_REP USDT_ZEC -3.932324
## 11 USDT_ETH USDT_ZEC -3.800008
## 12 USDT_ZEC USDT_REP -3.707490
## 13 USDT_REP USDT_DASH -3.707416
## 14 USDT_REP USDT_ETH -3.678356
## 15 USDT_REP USDT_XMR -3.618069
## 16 USDT_DASH USDT_REP -3.599772
## 17 USDT_XMR USDT_REP -3.575737
## 18 USDT_DASH USDT_ETH -3.262312
## 19 USDT_ETH USDT_REP -3.086158
9. Cross Validation April 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-04-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 31 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_REP USDT_DASH -5.521194
## 2 USDT_DASH USDT_REP -5.259461
## 3 USDT_XMR USDT_ETH -5.000186
## 4 USDT_XMR USDT_DASH -4.763988
## 5 USDT_REP USDT_LTC -4.739366
## 6 USDT_XMR USDT_BTC -4.179513
## 7 USDT_XMR USDT_REP -4.145031
## 8 USDT_XMR USDT_LTC -4.096683
## 9 USDT_ETH USDT_XMR -4.092585
## 10 USDT_ETH USDT_DASH -4.072278
## # ... with 21 more rows
10. Cross Validation March 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-03-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 11 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_LTC USDT_BTC -5.956905
## 2 USDT_LTC USDT_DASH -5.402013
## 3 USDT_LTC USDT_ZEC -5.369503
## 4 USDT_LTC USDT_REP -5.335098
## 5 USDT_LTC USDT_XMR -5.232308
## 6 USDT_LTC USDT_ETH -5.213867
## 7 USDT_XMR USDT_BTC -3.461436
## 8 USDT_REP USDT_XMR -3.190210
## 9 USDT_XMR USDT_REP -3.171302
## 10 USDT_XMR USDT_DASH -3.151879
## 11 USDT_BTC USDT_LTC -3.082816
11. Cross Validation February 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-02-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 28 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_DASH USDT_XMR -4.798816
## 2 USDT_XMR USDT_DASH -4.766860
## 3 USDT_ETH USDT_BTC -4.739089
## 4 USDT_XMR USDT_BTC -4.501784
## 5 USDT_BTC USDT_ETH -4.425492
## 6 USDT_BTC USDT_XMR -4.420176
## 7 USDT_REP USDT_ETH -4.250730
## 8 USDT_DASH USDT_BTC -4.147420
## 9 USDT_ETH USDT_REP -4.040995
## 10 USDT_BTC USDT_DASH -4.031222
## # ... with 18 more rows
12. Cross Validation January 2017
plot_many(pricing_data = pricing_data,
time_resolution = time_resolution,
cutoff_date = "2017-01-01",
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type,
number_pairs = number_pairs)## # A tibble: 13 x 3
## coin_y coin_x adf_stat
## <chr> <chr> <dbl>
## 1 USDT_ZEC USDT_BTC -4.288920
## 2 USDT_DASH USDT_REP -4.266037
## 3 USDT_REP USDT_DASH -3.978244
## 4 USDT_BTC USDT_ZEC -3.805267
## 5 USDT_ZEC USDT_LTC -3.791873
## 6 USDT_ETH USDT_DASH -3.784793
## 7 USDT_REP USDT_XMR -3.723175
## 8 USDT_DASH USDT_ETH -3.445868
## 9 USDT_LTC USDT_ZEC -3.424585
## 10 USDT_XMR USDT_REP -3.417472
## 11 USDT_DASH USDT_XMR -3.149624
## 12 USDT_ZEC USDT_XMR -3.120132
## 13 USDT_ETH USDT_REP -3.114262
13. Cross Validation Full
results <- backtest_strategy_full(pricing_data = pricing_data,
time_resolution = time_resolution,
train_window = train_window,
test_window = test_window,
quote_currency = quote_currency,
adf_threshold = adf_threshold,
rolling_window = rolling_window,
stop_threshold = stop_threshold,
signal_logic = signal_logic,
model_type = model_type) ## [1] "Cross validating strategy."
## [1] "Using train set from 2016-12-16 to 2017-01-01."
## [1] "Using test set from 2017-01-01 to 2017-01-11."
## [1] "Cross validating strategy."
## [1] "Using train set from 2016-12-26 to 2017-01-11."
## [1] "Using test set from 2017-01-11 to 2017-01-21."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-01-05 to 2017-01-21."
## [1] "Using test set from 2017-01-21 to 2017-01-31."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-01-15 to 2017-01-31."
## [1] "Using test set from 2017-01-31 to 2017-02-10."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-01-25 to 2017-02-10."
## [1] "Using test set from 2017-02-10 to 2017-02-20."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-02-04 to 2017-02-20."
## [1] "Using test set from 2017-02-20 to 2017-03-02."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-02-14 to 2017-03-02."
## [1] "Using test set from 2017-03-02 to 2017-03-12."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-02-24 to 2017-03-12."
## [1] "Using test set from 2017-03-12 to 2017-03-22."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-03-06 to 2017-03-22."
## [1] "Using test set from 2017-03-22 to 2017-04-01."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-03-16 to 2017-04-01."
## [1] "Using test set from 2017-04-01 to 2017-04-11."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-03-26 to 2017-04-11."
## [1] "Using test set from 2017-04-11 to 2017-04-21."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-04-05 to 2017-04-21."
## [1] "Using test set from 2017-04-21 to 2017-05-01."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-04-15 to 2017-05-01."
## [1] "Using test set from 2017-05-01 to 2017-05-11."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-04-25 to 2017-05-11."
## [1] "Using test set from 2017-05-11 to 2017-05-21."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-05-05 to 2017-05-21."
## [1] "Using test set from 2017-05-21 to 2017-05-31."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-05-15 to 2017-05-31."
## [1] "Using test set from 2017-05-31 to 2017-06-10."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-05-25 to 2017-06-10."
## [1] "Using test set from 2017-06-10 to 2017-06-20."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-06-04 to 2017-06-20."
## [1] "Using test set from 2017-06-20 to 2017-06-30."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-06-14 to 2017-06-30."
## [1] "Using test set from 2017-06-30 to 2017-07-10."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-06-24 to 2017-07-10."
## [1] "Using test set from 2017-07-10 to 2017-07-20."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-07-04 to 2017-07-20."
## [1] "Using test set from 2017-07-20 to 2017-07-30."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-07-14 to 2017-07-30."
## [1] "Using test set from 2017-07-30 to 2017-08-09."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-07-24 to 2017-08-09."
## [1] "Using test set from 2017-08-09 to 2017-08-19."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-08-03 to 2017-08-19."
## [1] "Using test set from 2017-08-19 to 2017-08-29."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-08-13 to 2017-08-29."
## [1] "Using test set from 2017-08-29 to 2017-09-08."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-08-23 to 2017-09-08."
## [1] "Using test set from 2017-09-08 to 2017-09-18."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-09-02 to 2017-09-18."
## [1] "Using test set from 2017-09-18 to 2017-09-28."
## [1] "Cross validating strategy."
## [1] "Using train set from 2017-09-12 to 2017-09-28."
## [1] "Using test set from 2017-09-28 to 2017-10-08."ggplot(results, aes(x = date_time)) +
geom_line(aes(y = return_strategy_cumulative), colour = "blue", size = 1) +
geom_hline(yintercept = 1, colour = "black") +
labs(title = "Strategy Return vs Buy Hold Return", x = "Date", y = "Cumulative Return") 
print(results[["return_strategy_cumulative"]][nrow(results)]) ## [1] 4.683395